Parameter Estimation For Artesian Aquifer In Table Mountain Group Aquifers By Analytical Method

Accurate parameter estimation for fractured-rock aquifers is very challenging, due to the complexity of   fracture   connectivity,   particularly   when   it   comes   to   artesian   flow   systems   where   the potentiometric  is  above  the  ground  level,  such  as  semi-confined,  partially  confined  and  weak confined aquifers in Table Mountain Group (TMG) Aquifer. The parameter estimates of these types of aquifers are largely made through constant-head and recovery test methods. However, such tests are seldom carried out in the Table Mountain Group Aquifer in South Africa due to the lack of a proper testing unit made available for data capturing and an appropriate method for data interpretation. 

An artesian borehole of BH-1 drilled in TMG Peninsula Formation on the Gevonden farm in Western Cape Province was chosen as a case study. The potentiometric surface is above the ground level in the rainy season, while it drops to below ground level during the dry season. A special testing unit was designed and implemented in BH-1 to measure and record the flow rate during the free-flowing period, and the pressure changes during the recovery period. All the data were captured at a function of time for data interpretation at later stage. 

Curve-fitting software developed with VBA (Visual Basic Application) in Excel was adopted for parameter estimation based on the constant-head and recovery tests theories. The results indicate that a negative skin zone exists in the immediate vicinity of the artesian borehole in Rawsonville, and the  hydraulic  parameters  estimates  of  transmissivity  (T)  ranging  from  6.9  to  14.7 m2/d  and storativity  (S)  ranging  from  2.1×10-5   to  2.1×10-4   appear  to  be  reasonable  with  measured  data collected from early times. The effective radius is estimated to be 0.5 to 1.58 m. However, due to formation losses, the analytical method failed to interpret the data collected at later times. Consequently the analysed results by analytical solution with later stage data are less reliable for this case. Numerical modelling is proposed to address the issue in future.

Presenter Name
Xiaobin
Presenter Surname
Sun
Area
Western Cape
Conference year
2013